Note: Descriptions are shown in the official language in which they were submitted.
CA 02459111 2014-03-27
DIRECT ADDITION OF BETA-AIVIINOENONES IN ORGANIC
ELECTROLYTES OF NONAQUEOUS CELLS EMPLOYING SOLID
CATHODES
DESCRIPTION
Technical Field
The invention relates to a method for manufacturing a nonaqueous cell
employing an
anode, a cathode comprising an active cathode material and an organic
electrolyte. The method
involves the direct addition of an additive such as beta-aminoenone,
preferably 4-amino-3-penten-
2-one, to the electrolyte to aid in reducing any undesirably high initial open
circuit voltage such
as that normally observed with cathodes such as FeS2.
Background Art
The development of high energy cell systems requires the compatibility of an
electrolyte possessing desirable electrochemical properties with highly active
anode
materials, such as lithium, calcium, sodium and the like, and the efficient
use of high energy
density cathode materials. Such suitable high energy density cathode materials
includes
iron sulfides such as FeS2 and FeS, carbon flourides such as CF,-, metal
oxides such as V205,
W03, M003, lead oxides such as Pb304, Pb02 and Pb0, cobalt oxides such as
Co304,
manganese oxides such as Mn02, In2S3, NiS, metal chromates such as Ag2Cr04,
metal
phosphates such as Ag3PO4, LiCo02, LiMn204, Bi,03, CuO and Cu20 and metal
sulfates
such as CuSO4. The use of aqueous electrolytes is precluded in these systems
since the
anode materials are sufficiently active to react with water chemically.
Therefore, in order to
realize the high energy density obtainable through use of these highly
reactive anodes and
high energy density cathodes, it is necessary to use a nonaqueous electrolyte
system.
CA 02459111 2014-03-27
Many cell or battery applications, particularly in transistorized devices,
such as
hearing aids, watches, calculators, and the like, require a substantially
unipotential discharge
source for proper operation. However, it has been found that in many
nonaqueous cells
employing positive active materials, which include conductive additives such
as graphite
andJor carbon, the cell upon initially being discharged, exhibits a high
voltage whereupon
the cell then proceeds to reach its lower operating discharge voltage level
only after a
certain time period has elapsed. The time period for the cell to reach its
intended operating
discharge voltage level usually depends on the discharge rate through the load
and thus,
depending on the apparatus it is to power, could result in a period extending
up to several
hours or .even days. This phenomenon has serious Ch-awbacks when a cell is
intended to be
used in electronic devices requiring a substantially unipotential discharge
source for proper
operation. In some of these electronic devices, any initial voltage peak
substantially
exceeding the intended operating voltage for the device could result in
serious damage to
the electronic components of the device.
One approach to protect devices from batteries exhibiting high voltages prior
to
leveling off to their desired operating voltage level is to add additional
electronic circuit
components to protect the main operating components of the device. However,
this not only
adds to the expense of the device but also would result in enlarging the
device to
accommodate the protective circuitry. Larger devices run counter to the
current emphasis
placed on miniaturization, where it has become necessary for the battery
industry to design
smaller and smaller electrochemically powered cells.
Another approach designed to reduce an initially high and potentially
disadvantageous voltage is to subject such cells to a pre-discharge regimen
prior to initial
use by an end user. - But such pre-discharge regimens are time consuming and
costly, and
2
CA 02459111 2014-03-27
reduce the capacity of the cell. Therefore, it is desirable to reduce or
eliminate any required
pre-discharge.
It has been suggested in U.S. Pat. No. 4489144 that the addition of an
isoxazole
derivative, such as 3,5-dimethylisoxazole, to the electrolyte may compensate
for or eliminate
this initial high voltage during discharge of nonaqueous cells. It was
believed that the
isoxazole derivative reacted with unwanted species that presumably caused the
initial high
voltage by reducing these species. The entire disclosure of U.S. Pat. No.
4489144 may be
referred to for further details.
It has now been determined that a favorable reduction in initial open circuit
voltage in
a nonaqueous cell can be achieved by the addition of an additive as defined
below to the
electrolyte. Accordingly, it is an aspect of the within invention to provide a
method for
manufacturing a cell comprising a highly active anode such as lithium, a
cathode comprising
an active material such as FeS2, and a nonaqueous electrolyte comprising an
additive, such as
4-amino-3-penten-2-one.
In a broad aspect the invention comprehends a method for manufacturing a
nonaqueous
cell comprising the steps of providing an anode, an organic electrolyte
solution comprising a
solute dissolved in an organic solvent, and a solid cathode, and adding an
additive to the
electrolyte solution, the additive added to the electrolyte solution in a
range of 0.1 to 5.0
volume percent based on the volume of the organic solvent and defined as any
one of:
-CO-CR=C(N112)-
where R is hydrogen or an alkyl group; or
-CO-CR=C(NR2R 3)
where R, R2 and R3 are hydrogen or an alkyl group; or related azirenes of the
formula
-CO-CRR'-C(NR2R3)(R")-
where R, R', R", R2 and R3 are hydrogen or an alkyl group.
3
CA 02459111 2014-03-27
Detailed Description Of The Invention
The addition of the additive, such as 4-amino-3-penten-2-one, to nonaqueous
organic
electrolytes is effective in reducing and stabilizing the undesirably high
initial open circuit
voltages observed with cathode materials such as FeS2. As used herein, the
term "additive"
will be used to describe a material containing any of the above structures.
A preferred additive is a beta-aminoenone such as 4-amino-3-penten-2-one.
The additive is preferably employed in the concentration range of between 0.1
and 5.0
volume percent based on the volume of the electrolyte solvent, preferably
between 0.2 and 4.0
volume percent, and still more preferably between 0.2 and 2.0 volume percent,
for most
nonaqueous electrochemical couples.
The effectiveness of such additive in controlling initially high open circuit
voltage was
first discovered while examining the electrolyte solution of nonaqueous cells
employing carbon
and/or graphite-containing FeS2 cathodes and lithium anodes. These cells were
manufactured
with an electrolyte solvent that included 3,5-dimethylisoxazole (DMI), an
isoxazole derivative,
as described in U.S. Pat. No. 4489144. Upon aging, gas chromatography (GC)
studies on the
electrolyte from the undischarged cells revealed that the DMI was no longer
detectable.
Instead, an amine was detected which was not initially present in the
electrolyte. The amine
was identified as 4-amino-3-penten-2-one (AP), one member of a class of beta-
aminoenones.
A second compound was also detected which was not initially present in the
electrolyte: 2,4-
pentanedione (PD). It is believed that the presence of PD is at least in part
an artifact
produced by exposure of the electrolyte to water during the GC sample
preparation.
4
CA 02459111 2004-02-27
WO 03/030279 PCT/US02/31048
The reaction mechanism involving DMI and AP was further investigated by GC
analysis of electrolytes from vial storage studies. In these studies, various
components of a
Li/FeS2 cell were isolated and combined with electrolyte containing DMI. The
procedure in
general was as follows: A 3/4 inch square piece of cell component sample was
placed in a 1
ounce jar. An electrolyte was prepared containing 1 mole of LiCF3S03 added to
1 liter of
25:75:0.2 DIOX:DME:DMI by volume, where DIOX is 1,3-clioxolane and DME is 1,2-
dimethoxyethane. One hundred micro liters of electrolyte was added to a vial.
The vial was
then capped and stored in the dry box for three hours. At the end of this
period, the
electrolyte was recovered in methylene chloride, the solution was extracted
with water and
the methylene chloride layer was analyzed.
Initially, electrolyte reaction with the lithium, separator material and
cathode stock
was examined. The results are presented in Table 1.
TABLE 1:
PERCENT OR PPM BY GC PEAK AREA
REACTANTS % DIOX % DME % DMI ppm AP
Electrolyte + 23.42 76.14 .44 ND*
lithium foil
Electrolyte + 23.66 75.91 .44 ND*
separator
Electrolyte + 21.97 77.87 .09 614
cathode stock
Electrolyte 25.48 74.14 .38 ND*
*None detected.
The above data indicate that DMI is degraded in a reaction involving the
cathode. To
further isolate the point of DMI degradation, cathode mix was isolated from
the aluminum
foil substrate and the vial test was run on the cathode mix and this aluminum
separately.
The mix and the aluminum substrate were recombined and another vial test was
run. A
CA 02459111 2004-02-27
WO 03/030279
PCT/US02/31048
further vial test was run on cathode mix and fresh aluminum. The results are
presented in
Table 2.
TABLE 2:
PERCENT OR PPM BY GC PEAK AREA
REACTANTS % DIOX % DME % DMI ppm AP ppm PD
Electrolyte + 23.88 75.52 .44 ND* ND*
cathode mix
Electrolyte + 24.04 75.46 .43 ND* ND*
alum. substrate
Electrolyte + 23.61 76.06 .25 362 534
cathode mix +
alum. substrate
Electrolyte + 24.18 75.45 .31 47 477
cathode mix +
fresh aluminum
*None detected.
In a further attempt to isolate the reactive species in the degradation of DMI
and the
generation of AP, cathode mix components were individually subjected to the
vial test in
combination with fresh substrate aluminum. These results are presented in
Table 3.
TABLE 3:
PERCENT OR PPM BY GC PEAK AREA
REACTANTS % DIOX % DME % DMI ppm AP ppm PD
Electrolyte + FeS2 20.27 78.90 .42 ND* ND*
(undried)
Electrolyte + Acetylene 25.24 74.38 .38 ND* ND*
black
Electrolyte + polyethylene 25.32 74.27 .41 ND* ND*
/polypropylene copolymer
Electrolyte + polyethylene 25.13 74.47 .38 ND* ND*
oxide
*None detected.
6
CA 02459111 2004-02-27
WO 03/030279 PCT/US02/31048
The above data indicates that the DMI degradation reaction to produce AP only
takes place
in the presence of both the cathode mix and an aluminum substrate current
carrier, and
suggests that the degradation reaction involves a possible interaction of the
cathode mix
with the aluminum cathode current carrier.
Further experiments were carried out to determine whether the direct addition
of AP
or PD into the electrolyte solvent in place of DMI would show similar
advantageous results
in reducing the open circuit voltage of an undischarged cell. Several lots of
nonaqueous AA
size lithium anode iron pyrite cathode cells were constructed using the
electrolyte described
above with the following variations: a) no DMI, b) 0.2% DMI, c) 0.2% PD
(instead of
DMI) and d) 0.2% AP (instead of DMI), all by volume percent of the electrolyte
solvent.
The pyrite for these cells came from different production batches as indicated
in the attached
tables but was essentially the same material. The cells were stored either at
ambient
temperature (21 C) or at 60 C. The cells were removed from storage at the
indicated period
in the tables below and the open circuit voltage for the cell was measured.
The cells were
then autopsied and GC analysis was performed on the electrolyte. The results
for the
ambient cells are presented in Table 4 and for the 60 C cells in Table 5.
Significantly, none
of these cells contained any detectable amounts of DMI even after only two
weeks at
ambient temperature, indicating that the degradation reaction is completed
relatively
quickly, even at ambient storage temperatures.
TABLE 4: STORAGE AT 21 C
Electrolyte description/ Pyrite lot # OCV prior to ppm AP (by GC ppm PD (by
GC
storage duration autopsy (Volts) peak area) peak area)
No DMI 171
2 weeks 1.727 ND* ND*
7 weeks 1.757 ND* ND*
23 weeks 1.785 ND* ND*
No DMI 172
7
CA 02459111 2004-02-27
WO 03/030279 PCT/US02/31048
Electrolyte description/ Pyrite lot # OCV prior to ppm
AP (by GC ppm PD (by GC
storage duration autopsy (Volts) peak area) peak area)
2 weeks 1.732 ND* ND*
10 weeks 1.760 ND* ND*
16 weeks 1.775 ND* ND*
0.2% DMI 171
2 weeks 1.725 1441 ND*
7 weeks 1.748 334 ND*
,
23 weeks 1.748 ND* ND*
0.2%DMI 172
2 weeks 1.726 1080 ND*
10 weeks 1.730 51 ND*
16 weeks 1.745 32 ND*
0.2%PD 171
2 weeks 1.725 ND* 39
7 weeks 1.758 ND* ND*
23 weeks 1.780 ND* ND*
0.2%AP (supplied by 172
Lancaster Synthesis)
2 weeks 1.727 1379 ND*
10 weeks 1.730 ND* ND*
16 weeks 1.740 304 ND*
0.2%AP (supplied by 172
TCI)
2 weeks 1.727 2215 ND*
10 weeks 1.740 192 ND*
16 weeks 1.740 ND* ND*
*None detected.
TABLE 5: STORAGE AT 60 C
Electrolyte description/ Pyrite lot # OCV PRIOR ppm AP
(by GC ppm PD (by GC
storage duration TO AUTOPSY peak area) peak area)
(Volts)
No DMI 171
4 weeks 1.846 ND* ND*
12 weeks 1.847 ND* ND*
No DMI 172
4 weeks 1.830 ND* ND*
12 weeks 1.830 ND* ND*
.2% DMI 171
4 weeks 1.798 ND* ND*
12 weeks 1.837 ND* ND*
'
.2%DMI 172
4 weeks 1.810 ND* ND*
12 weeks 1.824 ND* ND*
.2%PD 171
8
'
CA 02459111 2014-03-27
=
Electrolyte description/ Pyrite lot # OCV PRIOR ppm AP (by
GC ppm PD (by GC
storage duration TO AUTOPSY peak area) peak area)
(Volts)
4 weeks 1.840 ND* 32
12 weeks 1.834 ND* ND*
.2%AP (supplied by 172
Lancaster Synthesis)
4 weeks 1.810 ND* ND*
12 weeks 1.750 ND* ND*
.2%AP (supplied by 172
TCI)
4 weeks 1.760 267 ND*
12 weeks 1.790 ND* ND*
*None detected
A study of the open circuit voltages for the cells in Table 4 reveals the
following: at
16 weeks ambient, the cells with DMI in the electrolyte have a 30 mV lower
open circuit
voltage than the cells without DMI, confirming the observation disclosed in US
Pat. No.
=
4489144 that DMI aids in reducing undesirably high open circuit voltage in
such cells. The
cells containing AP have a 35 mV lower open circuit voltage than the cells
with no DMI.
On the other hand, at 23 weeks ambient there is a 37 mV difference between the
DMI cells
and the no added DMI cells, but only a 5 mV difference between the PD cells
and the cells
without DMI. In general, it can be concluded that the cells with AP added to
the electrolyte
behave in the same manner as do cells with added DMI in reducing open circuit
voltage,
while the cells with PD added to the electrolyte behave in the same manner as
do cells
without any DMI added to the electrolyte. The cells represented in Table 5
demonstrate that
the addition of AP to the electrolyte reduces open circuit voltage.
Literature studies show that AP can be obtained by the partial reduction of
DMI.
However, it is unexpected that DMI would be reduced in a reaction with the
cathode, in
view of the presence of an even stronger reducing agent, the lithium metal
anode. However,
when DMI is reduced with butyl lithium (often used to mimic the reducing power
of lithium
9
CA 02459111 2014-03-27
metal), the reaction product is 3-methyl-5-hydroxymethyl isoxazole, a compound
that is not
a beta-aminoenone as defined herein. The fact that this reaction product is
not seen in the
studies represented in Tables 1 through 5, together with the fact that AP does
react readily
with lithium metal, further supports the proposition that DMI is reduced in a
reaction with
the cathode to lower the cathode voltage and the resulting AP is consumed by
the lithium
anode. Based on this information, it would not be expected that AP would react
with the
cathode to lower its voltage. Lowering the cathode voltage presumably involves
reducing
the cathode. Thus, the more reducing power the additive has the more likely it
is to lower
the cathode voltage. AP, on the other hand, is a reduced form of DIVII in that
AP has two
additional hydrogen atoms. AP, being already partly reduced, has less reducing
power than
DML Consequently, one would not have expected AP to be as effective as DM1 or
even
effective at all in lowering the cathode voltage. However, the results in
Tables 1 through 5
demonstrate that AP behaves in the same manner as does D1\4I and does lower
the cathode
voltage.
The amount of the additive below 0.1 volume percent based on the volume
of the electrolyte solvent would likely not provide sufficient feducing
material to quickly
and effectively reduce impurities and/or unwanted active species in the cell
system. An
amount above 5.0 volume percent based on the volume of the electrolyte solvent
may
provide excessive reducing material)that could cause deleterious effects on
other desired
aspects of the cell. Although one embodiment of the invention, if focusing
solely on
voltage reduction, would include the use in the cathode of a metallic reducing
agent such as
zinc in conjunction with the additive in the electrolyte, the additive can be
used without the metallic reducing agent in the cell, without departing from
the scope of the
within invention. The omission of such a reducing agent may in fact be
preferable, since
CA 02459111 2014-03-27
such reducing agents may produce other undesirable features within the cell.
If the within invention is practiced in conjunction with the addition of a
metallic
reducing agent, the metallic reducing agent can either be employed in a
mixture in the
forming of a cathode or it could be placed in contact with the cathode. Thus,
any discrete
material such as a layer, a coating, a metallic screen, a metallic strip, a
porous metallic strip,
or a powder and could be employed as long as it is in electronic and ionic
contact with the
cathode. The choice of a metallic reducing agent will depend upon the voltage
of the cell's
active cathode material with reference to the anode of the cell. For example,
while zinc may
be satisfactory for a nonaqueous lithium/FeS2 system, it may not be suitable
for a
nonaqueous lithium/Ag90 system since, in the latter, a less anodic (negative)
metallic
reducing agent may be required, such as tin or lead. Thus, once the EMF series
is
determined for a specific nonaqueous electrolyte and anode/cathode system, one
can select
which metallic reducing agents can be used in conjunction with the additive.
In the nonaqueous cell systems of this invention, suitable active cathode
materials
would include iron sulfides such as FeS2 and FeS, carbon flourides such as CF,
metal
oxides such as V205, W03, Mo03, lead oxides such as P11304, Pb02 and Pb0,
cobalt oxides
such as Co304, manganese oxides such as Mn02, In2S3, NiS, metal chromates such
as
Ag2CrO4, metal phosphates such as Ag3PO4, LiCo02, UM:1204, B1203, CUO and Cu20
and
metal sulfates such as CuSO4. Highly active anodes for use in nonaqueous
systems
according to this invention would be consumable metals and include aluminum,
the alkali
metals, alkaline earth metals and alloys of alkali metals or alkaline earth
metals with each
other and other metals. The term "alloys" as used herein is intended to
include mixtures,
solid solutions, such as lithium-magnesium, and the intermetallic compounds,
such as
11
CA 02459111 2007-12-14
lithium monoahaminide. The preferred anode materials are lithium, sodium,
potassium,
calcium, magnesium and alloys thereof. Carbon anodes such as are used in
lithium ion cells
are also suitable anodes for use with the present invention. A preferred anode
material to be
used in conjunction with an FeS2 cathode having an aluminum cathode substrate
would be a
lithium-aluminum alloy such as that disclosed in US Pat. No. 5514491, which
may be
referred to for further details. One such alloy is 0.5% aluminum by weight and
is
available from Chemetall-Foote.
In nonaqueous cells, useful organic solvents employed alone or mixed in the
cell of
this invention include the following classes of compounds: alkylene nitriles:
e.g.,
crotonitrile (liquid ranges, -51.1 to 120 C); trialkyl borates: e.g.,
trimethyl borate, (CH30)3B
(liquid range, -29.3 C to 67 C); tetraalkyl silicates: e.g., tetramethyl
silicate, (CH30)4Si
(boiling point, 121 C); nitroalkanes: e.g., nitromethane, CH3NO2 (liquid
range, -17 C to
100.8 C); alkylnitriles: e.g., acetonitrile, CH3CN (liquid range, -45 C to
81.6 C);
dialkylamides: e.g., dimethylformamide, HCON(CH3)2 (liquid range, -60.48 C to
149 C);
lactams: e.g., N-methylpyrrolidone, (liquid range, -16 C to 202 C);
monocarboxylic acid
esters: e.g., ethyl acetate (liquid range, -83.6 C to 77.06 C); orthoesters:
e.g.,
trimethylorthofoimate, HC(OCH3)3 (boiling point, 103 C); lactones: e.g.,
(gamma)butyrolactone, (liquid range, -42 C to 206 C); dialkyl carbonates:
e.g., dimethyl-
carbonate, OC(OCH3)2 (liquid range, 2 C to 90 C); alkylene carbonates: e.g.,
propylene
carbonate (liquid range, -48 C to 242 C), ethylene carbonate and vinylene
carbonate;
monoethers: e.g., diethyl ether (liquid range, -116 C to 34.5 C); polyethers:
e.g., 1,1-and
1,2-dimethoxyethane (liquid ranges, -113.2 C to 64.5 C and -58 C to 83 C,
respectively);
cyclic ethers: e.g., tetrahydpfuran (liquid range, -65 C to 67 C), 1,3-
dioxolane (liquid
range, -95 C to 78 C) and substituted dioxolanes; nifroaromatics: e.g.,
nitrobenzene (liquid
12
CA 02459111 2004-02-27
WO 03/030279 PCT/US02/31048
range, 5.7 C to 210.8 C); cyclic sulfones: e.g., sulfolane, (melting point, 22
C), 3-
methylsulfolane (melting point, -1 C); saturated heterocyclics: e.g.,
tetrahydrothiophene
(liquid range, -96 C to 121 C); 3-methyl-2-oxazolidone (melting point, 15.9
C); five-
membered unsaturated heterocyclics: e.g. 1-meth-ylpyrrole (boiling point, 114
C), 2,4-
dimethylthiazole (boiling point, 144 C), and furan (liquid range, -85.65 C to
31.36 C);
dialkyl sulfoxides: e.g., dimethyl sulfoxide (liquid range, 18.4 C to 189 C);
dialkyl sulfates:
e.g., dimethylsulfate (liquid range, -31.75 C to 188.5 C); dialkyl sulfites:
e.g.,
dimethylsulfite (boiling point, 126 C); alkylene sulfites: e.g., ethylene
glycol sulfite (liquid
range, -11 C to 173 C); halogenated alkanes: e.g., methylene chloride (liquid
range, -95 C
to 40 C); 1,3-dichloropropane (liquid range, -99.5 C to 120.4 C). Of the
above, the
preferred solvents are sulfolane; tetrahydrofuran; methyl-substituted
tetrahydrofuran; 1,3-
dioxolane; alkyl substituted 1,3-dioxolane; 3-methyl-2-oxazolidone; propylene
or ethylene
carbonate; (gamma)butyrolactone; ethylene glycol sulfite; dimethylsulfite;
dimethyl
sulfoxide; and 1,1-, 1,2-dimethoxyethane and glymes. Of the preferred
solvents, the best are
3-methyl-2-oxazolidone, propylene or ethylene carbonate, 1,2-dimethoxyethane
and 1,3-
dioxolane because they appear more chemically inert to battery components and
have wide
liquid ranges, and especially because they permit highly efficient utilization
of the cathode
materials.
The ionizing solute for use in the invention may be a simple or double salt or
mixtures thereof, e.g., LiCF3S03 or LiC104 or LiI, which will produce an
ionically
conductive solution when dissolved in one or more solvents. Useful solutes
include
complexes of inorganic or organic Lewis acids and inorganic ionizable salts.
The only
requirements for utility are that the salts, whether simple or complex, be
compatible with the
solvent or solvents being employed and that they yield a solution which is
sufficiently
13
CA 02459111 2004-02-27
WO 03/030279
PCT/US02/31048
ionically conductive. According to the Lewis or electronic concept of acids
and bases, many
substances, which contain no active hydrogen can act as acids or acceptors of
electron
doublets. The basic concept is set forth in the chemical literature (Journal
of the Franklin
Institute, Vol. 226, July/December 1938, pages 293-313 by G. N. Lewis). A
suggested
reaction mechanism for the manner in which these complexes function in a
solvent is
described in detail in U.S. Pat. No. 3,542,602 wherein it is suggested that
the complex or
double salt formed between the Lewis acid and the ionizable salt yields an
entity which is
more stable than either of the components alone. Typical Lewis acids suitable
for use in
the present invention include aluminum fluoride, aluminum bromide, aluminum
chloride,
antimony pentachloride, zirconium tetrachloride, phosphorus pentachloride,
boron fluoride,
boron chloride and boron bromide. Ionizable salts useful in combination with
the Lewis
acids include lithium fluoride, lithium chloride, lithium bromide, lithium
sulfide, sodium
fluoride, sodium chloride, lithium bromide, potassium fluoride, potassium
chloride and
potassium bromide.
The within invention can be utilized in the manufacture of both primary and
secondary nonaqueous cells in a wide variety of cell constructions, including
but not limited
to cylindrical wound cells, cylindrical bobbin cells, miniature button cells,
cells with one or
more planar electrodes arranged in a co-facial orientation or in a co-planar
orientation and
prismatic cells. Further, although the experiments were performed by adding AP
to the
electrolyte, one of skill in the art will appreciate that the same result can
be expected by the
addition of AP to the cathode of the cell, without departing from the scope of
the within
invention.
14
